Dear STATA users,

Although it feels like a simple question, I have spent a lot of time reading STATA's vec guide and other resources and I cannot figure this out on my own.

I am trying to estimate a VECM to see how several (six) economic fundamentals influence the long-run equilibrium exchange rate of a currency.
Additionally, I want to compare different proxies of these fundamentals to see which one can explain movements in the exchange rate best.

When I use the vecrank command, I find there are six cointegrated relationships.
However, I am only interested in explaining movements in the exchange rate. Does that mean I can run the vec command with rank one?

I also tried specifyin "rank(6)", in case I do need to specify rank(6).
But this gives me an output table in which all coefficients are zero, except for the last one. If I specify rank(5), all coefficients are zero except for the last two.

I would like to know if it is OK to use rank one in this situation. If not, I would like to know how to interpret the outcomes I get when I use the rank(6) option.

Any help would be very much appreciated!

Kind regards,

Hero Hovinga